The midgut is the first organ in the mosquito that is invaded by malaria parasites. We sequenced 223 million reads from An. gambiae midgut cDNA libraries. In total, 22,889 unique midgut transcript models were generated from both An. gambiae strain sequences combined, and 76% are potentially novel. Of these novel transcripts, 50.5% mapped to regions between annotated genes and represent novel intergenic transcripts (NITs). Predicted models were validated for midgut expression using qRT-PCR and microarray analysis, and novel isoforms were confirmed by sequencing predicted intron-exon boundaries. This in-depth Illumina sequencing and assembly of the An. gambiae midgut transcriptome doubled the number of known transcripts and tripled the number of variants known in this mosquito species. It also revealed existence of a large number of lncRNA and opens new possibilities for investigating their biological function. This work was published in BMC Genomics The HPX2/NOX5 system potentiates NO toxicity and labels ookinetes as they traverse the midgut, making them visible to the mosquito complement-like system. We identify the JNK signaling pathway as a key regulator of midgut epithelial nitration by regulating expression of two key enzymes that potentiate nitration, heme peroxidase (HPX2) and NADPH oxidase 5 (NOX5). This pathway also regulates the basal levels of TEP1 and FBN9 expression in hemocytes. Disruption of the JNK signaling cascade prevents activation of the Anopheles gambiae complement-like immune system. The manuscript describing this work was published in Plos Pathogens. We recently found that the Pfs47 gene mediates Plasmodium evasion of the mosquito immune system. Infections with Pfs47 KO parasites are very low in the A. gambie G3 strain, that is very suscriptible to infection with NF54 wild-type (WT) parasites. We compared the effect of disrupting JNK signaling by silencing key genes from this pathway on infections with WT and Pfs47-KO parasites. We found that disrupting JNK signaling had no effect on WT infections, but greatly enhanced infection with KO parasites. A screen of genes differentially expressed in response to WT and KO parasites identified a caspase (S2) and a novel conserved gene of unknown function as key mediators of antiplasmodial responses to KO parasite. In depth pathway analysis revealed that midgut invasion by KO parasite triggers JNK singaling, activation of the AgL1 initiator Caspase (ortholog of Dm Dredd) and of several effector caspases that greatly reduce parasite survival. The presence of Pfs47 in WT parasite avoids activation of these responses and prevents detection of the parasite. In Africa the Pfs47 gene is highly polymorphic, but in other areas of the world it has a remarkable degree of geographic structure. We are testing the general hypothesis that different alleles of Pfs47 work better to evade the immune system of certain mosquito vectors. This hypothesis predicts that the different mosquito vectors in Africa will preferentially transmit parasites with certain Pfs47 alleles. We have a collaboration with Dr. Isabelle Morlais group in Cameroon. We are analyzing mosquito midguts infected with patient samples containing multiple parasite infections that were used to infect two mosquito different colonies (M and S form). We are analyzing the frequency of each of the Pfs47 present in the donor blood, the gametocytes in the samples and the oocysts that resulted from the infection. We will test the hypothesis that some Pfs47 haplotypes will be more successful in to infect M-form and others will do better in S-form mosquitoes, even when both mosquito forms were fed with the same donor blood. The entomology team in Mali is collecting mosquitoes from homes using the spray-catch method and doing ELISA assays to identify Plasmodium-falciparum infected mosquitoes. They are shipping the material from infected mosquitoes to NIH and we are extracting the genomic DNA, sequencing the Pfs47 gene. We identified 11 Pfs47 haplotypes in mosquitoes with a single infection and two most abundant represent 73% of the samples. The largest haplotype diversity was found in samples collected in October. With our current sample size, we did not find a significant difference in the frequency of Pfs47 haplotypes between M-form and S-form A. gambiae mosquitoes. We recently received the collections from 2012-2013 transmission season. Our priority for this coming year is to analyze Pfs47 circulating in A. funestus and compare them with those in A. gambiae, as these two mosquito species are more evolutionary divergent. At NIH, we have several colonies of anopheline mosquitoes that transmit malaria in different regions of the world (A. albimanus, A. gambiae M-form, A. dirus, A. stephensi). Recently, we also established an A. aquasalis colony, an important vector in Brazil. We are investigating the hypothesis that parasites expressing certain Pfs47 alleles ware be detected by the immune system of certain mosquito species, but are be invisible to other mosquito vectors. We have confirmed that the African NF54 line does not infect A. albimanus or A. aquasalis, and has low infectivity in A. dirus. Disruption of the complement-like system in A. albimanus results in a complete rescue of the infection, while there is a partial rescue of the infectivity in A. dirus. The Brazilian 7G8 line readily infects A. albimanus, but does not infect A. dirus. We also generated transgenic P. falciparum strains with the same genetic background (NF54) that only differ in the allelic variant of Pfs47 they express. We are currently genotyping these lines to establish whether changing the allele of Pfs47 expressed by a parasite is enough to evade the mosquito immune system and be transmitted by a new mosquito species. We found that adult mosquito females challenged with Plasmodium respond more efficiently to subsequent challenges and that the transfer of a hemocyte differentiation factor (HDF) in the hemolymph from challenged mosquitoes induces hemocyte differentiation and confers increased resistance to Plasmodium infection in the recipients. HDF appears to consist of a lipocalin that carries a bioactive lipid. RNAi-based silencing of this Apo-D-like abolished the release of HDF activity in the hemolymph extract, indicating that it is a critical component of the immune priming response. We have found that injection of synthetic lipid analogs recapitulate the phenotype observed in Plasmodium-infected mosquitoes and the response to HDF. We have developed a protocol to successfully deliver deuterium-labeled lipid precursors to determine the chemical nature of this mosquito eicosanoid and optimized our total lipid extraction protocol for lipidomics analysis. Hemocytes synthesize key components of the mosquito complement-like system, but their role in the activation of antiplasmodial responses has not been established. Activation of the Toll signaling in hemocytes greatly enhanced antiplasmodial immunity, indicating that hemocytes are active players in the activation of the complement-like system, through an effector/effectors regulated by the Toll pathway. Susceptible G3 mosquitoes responded with similar kinetics following infection with three different Plasmodium species, but the strength of the priming response was stronger in the less compatible mosquito-parasite pairs. The JNK, STAT and Toll pathway are required for hemocytes to respond to HDF, but the IMD pathway is dispensable. However, neither the Toll, JNK, STAT nor IMD pathways are required for HDF to be synthesized and released into the mosquito hemolymph. This work was published in the Journal of Innate Immunity.